Active Monitoring Method, System, Device and Medium for Suspicious Air Computing System

Abstract

The present invention relates to the field of air computing technology, and discloses an active listening method, system, device and medium for a suspicious air computing system. In the suspicious air computing system, a suspicious receiver collects illegal information through a suspicious sensor, and an active listener listens to and actively interferes with the suspicious receiver from collecting illegal information. The intelligent reflecting surface enhances the listening channel for the active listener to listen to the suspicious receiver collecting illegal information and weakens the suspicious channel for the suspicious receiver to collect illegal information; an optimization problem of minimizing the calculation mean square error of the active listener is constructed under the constraints of the calculation mean square error of the suspicious receiver, the transmission power of the suspicious sensor, the active interference power of the active listener and the phase matrix of the intelligent reflecting surface; the optimization problem is solved to obtain the target receiving power of the active listener, the target active interference power of the active listener and the target phase matrix of the intelligent reflecting surface, so as to realize the active listening of the suspicious air computing system.

Description

Technical Field

[0001] The present invention relates to the technical field of air computing, and particularly to an active listening method, system, device and medium for a suspicious air computing system. Background Art

[0002] Currently, a large number of devices, including sensors, coordinators, and drones, are deployed in various application scenarios of the Internet of Things (IoT), such as smart agriculture, smart transportation, and smart cities. Different from traditional cellular networks that only focus on the information of a single user, the IoT focuses on the information fusion of a large number of devices, such as the sum, average value, etc. of all sensor data. Air computing technology (AirComp) is considered an efficient information fusion technology that utilizes the superposition characteristics of multiple access channels to simultaneously collect and calculate all sensor data, achieving the goals of low latency and high spectral efficiency.

[0003] Wireless technology is a double-edged sword. While bringing convenience to people, it also brings new threats. Wireless technology can be used for both legal communication and illegal destruction by criminals or terrorists, threatening public safety. Therefore, in order to ensure people's safety and take timely intervention measures, it is extremely necessary to conduct legal monitoring on it. At present, legal monitoring can be roughly divided into two categories: passive monitoring and active monitoring. In passive monitoring, the monitor only eavesdrops on information, and the monitoring may fail when the monitor's channel is worse than that of the suspicious user; active monitoring uses a duplexer to simultaneously eavesdrop and send active interference to increase the eavesdropping success rate.

[0004] However, currently, it is only limited to the monitoring of traditional communication systems, and the monitoring of air computing systems is still in a missing stage. Air computing technology may also be illegally used. For example, an adversary can use sensors to collect battlefield information, and a commercial spy can use sensors to steal business secrets. Therefore, it is extremely necessary to monitor air computing systems. Summary of the Invention

[0005] The purpose of the present invention is to provide an active listening method, system, device and medium for a suspicious air computing system, which can realize the active listening of an air computing system.

[0006] To solve the above technical problems, an embodiment of the present invention provides an active listening method for a suspicious air computing system, which is applied to an active listening system. The suspicious air computing system includes a suspicious receiver and a suspicious sensor, and the active listening system includes an active monitor and an intelligent reflecting surface. There is a connection relationship between any two of the suspicious receiver, the suspicious sensor, the active monitor, and the intelligent reflecting surface;

[0007] Among them, the suspicious receiver is used to collect illegal information through a suspicious sensor, the active listener is used to listen to and actively interfere with the suspicious receiver's collection of illegal information, and there is self-interference in the active listener. The intelligent reflecting surface is used to enhance the listening channel for the active listener to listen to the suspicious receiver's collection of illegal information and weaken the suspicious channel for the suspicious receiver to collect illegal information;

[0008] The method includes:

[0009] According to the channel between the suspicious sensor and the suspicious receiver, the channel between the intelligent reflecting surface and the suspicious receiver, the channel between the suspicious sensor and the active listener, the channel between the suspicious sensor, the intelligent reflecting surface and the active listener, the channel between the active listener and the suspicious receiver, and the self-interference channel of the active listener, construct an active listening model for the suspicious air computing system;

[0010] According to the active listening model of the suspicious air computing system, through the transmission power of the suspicious sensor, the reception power of the suspicious receiver, and the phase matrix of the intelligent reflecting surface, obtain the computational mean square error when the suspicious receiver collects illegal information through the suspicious sensor;

[0011] According to the active listening model of the suspicious air computing system, through the reception power of the active listener, the active interference power of the active listener, and the phase matrix of the intelligent reflecting surface, obtain the computational mean square error when the active listener listens to the suspicious receiver's collection of illegal information;

[0012] Under the constraints of the computational mean square error of the suspicious receiver, the transmission power of the suspicious sensor, the active interference power of the active listener, and the phase matrix of the intelligent reflecting surface, construct an optimization problem with the goal of minimizing the computational mean square error of the active listener;

[0013] Solve the optimization problem with the goal of minimizing the computational mean square error of the active listener to obtain the target reception power of the active listener, the target active interference power of the active listener, and the target phase matrix of the intelligent reflecting surface;

[0014] Enable the active listener to listen to the suspicious receiver's collection of illegal information with the target reception power and actively interfere with the suspicious receiver's collection of illegal information with the target active interference power, and enable the intelligent reflecting surface to enhance the listening channel of the active listener and weaken the suspicious channel of the suspicious receiver with the target phase matrix, so as to actively listen to the suspicious receiver in the suspicious air computing system.

[0015] In some alternative embodiments, the solving of the optimization problem with the goal of minimizing the computational mean square error of the active listener includes:

[0016] Decompose the optimization problem aiming at minimizing the computational mean square error of the active listener, and obtain multiple sub-optimization problems that aim at minimizing the computational mean square error of the active listener and take the received powers of the suspicious receiver and the active listener, the transmit power of the suspicious sensor, the active interference power of the active listener, and the phase matrix of the intelligent reflecting surface as optimization variables respectively;

[0017] Solve each sub-optimization problem.

[0018] In some alternative embodiments, the solving each sub-optimization problem includes:

[0019] For each sub-optimization problem, use the alternating sequential rank-one constrained relaxation iteration SROCR algorithm to solve it.

[0020] In some alternative embodiments, the computational mean square error of the suspicious receiver when collecting illegal information through the suspicious sensor is expressed by the following formula:

[0021]

[0022] The computational mean square error of the active listener when listening to the suspicious receiver collecting illegal information is expressed by the following formula:

[0023]

[0024] Wherein, represents the signal misalignment error of the suspicious receiver, represents the error caused by the active interference of the active listener, represents the error caused by the noise of the suspicious receiver, represents the signal misalignment error of the active listener, represents the error caused by the full-duplex self-interference of the active listener, represents the error caused by the noise of the active listener, Φ = diag(φ1, φ2, …, φ N ) represents the phase matrix of the intelligent reflecting surface, n ∈ {1, 2, …, N}, θ n ∈ [0, 2π), represents the adjustment phase of the nth reflection unit of the intelligent reflecting surface, u D and u L respectively represent the received powers of the suspicious receiver and the active listener, v k represents the transmit power of the kth suspicious sensor, w represents the active interference power of the active listener, and respectively represent the noise powers of the suspicious receiver and the active eavesdropper, and respectively represent the channels from the k-th suspicious sensor to the suspicious receiver, the intelligent reflecting surface, and the active eavesdropper, H ID and H IL respectively represent the channels from the intelligent reflecting surface to the suspicious receiver and the active eavesdropper, H LD represents the channel from the active eavesdropper to the suspicious receiver, H LL represents the full-duplex self-interference channel of the active eavesdropper.

[0025] In some alternative embodiments, the optimization problem aiming at minimizing the calculated mean square error of the active eavesdropper is represented by the following formula:

[0026]

[0027] In the formula, MSE L represents the calculated mean square error of the active eavesdropper, C1 represents the constraint condition of the calculated mean square error of the suspicious receiver, ε represents the preset covert constraint threshold, which is a constant, C2 represents the constraint condition of the transmission power of the suspicious sensor, P k represents the maximum transmission power of the k-th suspicious sensor, C3 represents the constraint condition of the active interference power of the active eavesdropper, P L represents the maximum transmission power of the active eavesdropper, C4 represents the constraint condition of the phase matrix of the intelligent reflecting surface, n ∈ {1, 2, …, N} represents the n-th reflecting unit of the intelligent reflecting surface.

[0028] In some alternative embodiments, before solving each sub-optimization problem, it further includes:

[0029] By setting the transmission power of the suspicious sensor, the active interference power of the active eavesdropper, and the phase matrix of the intelligent reflecting surface, the closed-form expressions of the optimal solutions of the received powers of the suspicious receiver and the active eavesdropper are obtained respectively, as shown in the following formulas:

[0030]

[0031]

[0032] In the formula, represents the equivalent channel from the k-th suspicious sensor to the suspicious receiver, represents the equivalent channel from the k-th suspicious sensor to the active eavesdropper;

[0033] By setting the received powers of the suspicious receiver and the active eavesdropper, the active interference power of the active eavesdropper, and the phase matrix of the intelligent reflecting surface, the closed-form expression of the optimal solution of the transmission power of the suspicious sensor is obtained by using the Lagrange multiplier method, as shown in the following formula:

[0034]

[0035] In the formula, represents the Lagrange multiplier corresponding to the constraint condition C2;

[0036] By given the received powers of the suspicious receiver and the active eavesdropper, the transmit power of the suspicious sensor, and the phase matrix of the intelligent reflecting surface, the closed-form expression of the optimal solution of the active interference power of the active eavesdropper is obtained, as shown in the following formula:

[0037]

[0038] In the formula, represents the eigenvector corresponding to the minimum eigenvalue of the matrix (A D / ε1) -1 A L ;

[0039] By given the received powers of the suspicious receiver and the active eavesdropper, the transmit power of the suspicious sensor, and the active interference power of the active eavesdropper, the closed-form expression of the optimal solution of the phase matrix of the intelligent reflecting surface is obtained, as shown in the following formula:

[0040]

[0041] In the formula, F = diag(f) represents a diagonal matrix with f as the diagonal elements, t represents an auxiliary variable,

[0042] c D = ([C D 1,1 , [C D 2,2 , …, [C D N,N ), c L = ([C L 1,1 , [C L 2,2 , …, [C L N,N ), [A] n,n represents the nth diagonal element of the matrix A,

[0043]

[0044] Obtained by performing eigenvalue decomposition on the optimal solution X opt of P8, g​​​​​​max is X opt The maximum eigenvalue λ max The corresponding eigenvector.

[0045] In some alternative embodiments, for each sub-optimization problem, the alternating successive rank-one constrained relaxation iteration SROCR algorithm is used for solution, including:

[0046] Set the initial value of the calculated mean square error of the active listener The initial value of the iteration number i = 1, the iteration accuracy δ, and the maximum iteration number I max , and set the initial values of the optimization variables; the optimization variables include the transmission power of the suspicious sensor The active interference power w of the active listener (0) And the initial value Φ of the phase matrix of the intelligent reflecting surface (0) ;

[0047] Based on the values of the optimization variables Φ in the previous iteration (i-1) , w (i-1) , according to the closed-form expressions of the optimal solutions of the received powers of the suspicious receiver and the active listener, respectively, find the optimal received power of the suspicious receiver And the optimal received power of the active listener

[0048] Based on the values of the optimization variables Φ in the previous iteration (i-1) , w (i-1) , According to the closed-form expression of the optimal solution of the transmission power of the suspicious sensor, find the optimal transmission power of the suspicious sensor

[0049] Based on the values of the optimization variables Φ in the previous iteration (i-1) , According to the closed-form expression of the optimal solution of the active interference power of the active listener, find the optimal active interference power w of the active listener (i) ;

[0050] Based on the values of the optimization variables in the previous iteration w (i) , according to the closed-form expression of the optimal solution of the phase matrix of the intelligent reflecting surface, find the optimal phase matrix Φ of the intelligent reflecting surface (i) ;

[0051] Increase the iteration number i = i + 1;

[0052] If i > I max Or Then stop the iteration.

[0053] An embodiment of the present invention further provides an active monitoring system for a suspicious air computing system. The suspicious air computing system includes a suspicious receiver and a suspicious sensor. The active monitoring system includes an active monitor and an intelligent reflecting surface, and there is a connection relationship between any two of the suspicious receiver, the suspicious sensor, the active monitor, and the intelligent reflecting surface;

[0054] Among them, the suspicious receiver is used to collect illegal information through the suspicious sensor. The active monitor is used to monitor and actively interfere with the suspicious receiver's collection of illegal information, and the active monitor has self-interference. The intelligent reflecting surface is used to enhance the monitoring channel for the active monitor to monitor the suspicious receiver's collection of illegal information and weaken the suspicious channel for the suspicious receiver to collect illegal information;

[0055] The active monitoring system further includes:

[0056] A monitoring model construction module, which is used to construct an active monitoring model of the suspicious air computing system according to the channels between the suspicious sensor and the suspicious receiver, the channels between the intelligent reflecting surface and the suspicious receiver, the channels between the suspicious sensor and the active monitor, the channels between the suspicious sensor, the intelligent reflecting surface and the active monitor, the channels between the active monitor and the suspicious receiver, and the self-interference channel of the active monitor;

[0057] A first data acquisition module, which is used to obtain the computational mean square error of the suspicious receiver when collecting illegal information through the suspicious sensor according to the active monitoring model of the suspicious air computing system, the transmission power of the suspicious sensor, the reception power of the suspicious receiver, and the phase matrix of the intelligent reflecting surface;

[0058] A second data acquisition module, which is used to obtain the computational mean square error of the active monitor when monitoring the suspicious receiver's collection of illegal information according to the active monitoring model of the suspicious air computing system, the reception power of the active monitor, the active interference power of the active monitor, and the phase matrix of the intelligent reflecting surface;

[0059] An optimization problem construction module, which is used to construct an optimization problem with the goal of minimizing the computational mean square error of the active monitor under the constraints of the computational mean square error of the suspicious receiver, the transmission power of the suspicious sensor, the active interference power of the active monitor, and the phase matrix of the intelligent reflecting surface;

[0060] An optimization problem solving module, which is used to solve the optimization problem with the goal of minimizing the computational mean square error of the active monitor, and obtain the target reception power of the active monitor, the target active interference power of the active monitor, and the target phase matrix of the intelligent reflecting surface;

[0061] An active listening module is used to enable an active listener to listen to a suspicious receiver for illegal information collection at a target receiving power, and actively interfere with the suspicious receiver for illegal information collection at a target active interference power, so that the intelligent reflecting surface enhances the listening channel of the active listener and weakens the suspicious channel of the suspicious receiver with a target phase matrix, so as to actively monitor the suspicious receiver in the suspicious air computing system.

[0062] An embodiment of the present invention also provides a computer device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, instructions executable by the at least one processor are stored in the memory, and the instructions are executed by the at least one processor to enable the at least one processor to execute the active listening method of the above-mentioned suspicious air computing system.

[0063] An embodiment of the present invention also provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the active listening method of the above-mentioned suspicious air computing system is implemented.

[0064] The active listening method of the suspicious air computing system provided by the present invention has at least the following beneficial effects:

[0065] By minimizing the computational mean square error when the active listener listens to and interferes with the suspicious receiver for illegal information collection, the success rate of active listening in the air computing system can be improved. Therefore, the present invention constructs an optimization problem with the goal of minimizing the mean square error of the active listener and solves it to obtain the operating parameters of each device in the system when the mean square error of the active listener is minimized, such as the receiving power of the active listener, the active interference power of the active listener, and the phase matrix of the intelligent reflecting surface, so that they can cooperate with each other with their respective operating parameters to achieve active listening of the suspicious air computing system.

[0066] It includes the operating parameters of the intelligent reflecting surface. The intelligent reflecting surface can be used to enhance the listening channel for the active listener to listen to and actively interfere with the suspicious receiver for illegal information collection and weaken the suspicious channel for the suspicious receiver to collect illegal information. Therefore, by setting an intelligent reflecting surface in the suspicious air computing system, the computational mean square error of the suspicious receiver when collecting illegal information through suspicious sensors can be increased, and on this basis, the computational mean square error when the active listener listens to and interferes with the suspicious receiver for illegal information collection is minimized, further improving the success rate of active listening. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] One or more embodiments are exemplarily illustrated by pictures in the corresponding drawings, and these exemplary illustrations do not limit the embodiments.

[0068] Figure 1Schematic diagram of a suspicious air computing system provided according to an embodiment of the present invention;

[0069] Figure 2 Flowchart of an active listening method for a suspicious air computing system provided according to an embodiment of the present invention;

[0070] Figure 3 Schematic diagram for comparing simulation results provided according to an embodiment of the present invention. Detailed implementation manners

[0071] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be described in detail below with reference to the accompanying drawings. However, those of ordinary skill in the art can understand that in the embodiments of the present invention, many technical details are provided to help readers better understand the present invention. However, even without these technical details and various changes and modifications based on the following embodiments, the technical solutions claimed by the present invention can still be implemented. The following division of each embodiment is for convenience of description and should not constitute any limitation to the specific implementation manner of the present invention. Each embodiment can be combined and cross-referenced with each other on the premise of not being contradictory.

[0072] The listening of the air computing system is different from that of the traditional communication system in the following aspects:

[0073] (1) Different research objects. The research object of the traditional communication system is a single user, while the research object of the air computing system is a large number of sensors.

[0074] (2) Different optimization objectives. The traditional communication system aims to maximize the security rate, while the air computing system aims to minimize the computational mean square error.

[0075] Therefore, the research on the listening of the air computing system is not a simple extension of the research on the listening of the traditional communication system. Based on this, in the present invention, for a suspicious air computing system, by transmitting active interference to the suspicious channel and using an Intelligent Reflecting Surface (IRS) to simultaneously enhance the listening channel and weaken the suspicious channel, the computational mean square error of the suspicious system is always greater than the threshold value, and the transceiver beam of the full-duplex listener and the phase of the intelligent reflecting surface are jointly optimized to minimize the listening computational mean square error.

[0076] An embodiment of the present invention relates to an active listening method for a suspicious air computing system, which is applied to an active listening system. The suspicious air computing system and the active listening system described in this embodiment are as Figure 1 shown. The suspicious air computing system includes: a suspicious receiver and suspicious sensors. The active listening system includes an active listener and an intelligent reflecting surface. Let the number of antennas of the suspicious receiver be MD , there are K suspicious sensors, and the number of transmitting and receiving antennas of the active eavesdropper are M T and M R , and the IRS has N reflecting elements. The intelligent reflecting surface is a metasurface composed of a large number of low-cost passive reflecting elements. Each element can independently adjust the amplitude and phase of the incident signal. Through the cooperation of each element, the intelligent reflecting surface can make the reflected signals coherently superpose at the receiving end and coherently cancel at the eavesdropping end, so as to achieve the purpose of ensuring system security and transmission performance simultaneously.

[0077] Among them, there is a connection relationship between any two of the suspicious receiver, suspicious sensors, active eavesdropper, and intelligent reflecting surface. The suspicious receiver is used to collect illegal information through the suspicious sensors, specifically by aggregating the illegal information collected by K suspicious sensors for air computing; the active eavesdropper is used to eavesdrop on and actively interfere with the suspicious receiver's collection of illegal information. The active eavesdropper specifically sends active interference while eavesdropping on the sensor information to reduce the success rate of suspicious link aggregation. The active eavesdropper has self-interference. In this embodiment, the active eavesdropper operates in full-duplex mode, so there is full-duplex self-interference; the intelligent reflecting surface is used to enhance the eavesdropping channel for the active eavesdropper to eavesdrop on the illegal information collected by the suspicious receiver and weaken the suspicious channel for the suspicious receiver to collect illegal information, so as to further improve the eavesdropping performance.

[0078] Next, the implementation details of the active eavesdropping method of the suspicious air computing system in this embodiment are specifically described. The following content is only the implementation details provided for easy understanding and is not necessary for implementing this solution. The specific process of the active eavesdropping method of the suspicious air computing system in this embodiment can be as Figure 2 shown, including:

[0079] Step 201, construct an active eavesdropping model of the suspicious air computing system according to the channels between the suspicious sensors and the suspicious receiver, the channels between the intelligent reflecting surface and the suspicious receiver, the channels between the suspicious sensors and the active eavesdropper, the channels between the suspicious sensors, the intelligent reflecting surface and the active eavesdropper, the channels between the active eavesdropper and the suspicious receiver, and the self-interference channel of the active eavesdropper.

[0080] Specifically, in this embodiment, based on the working principle among the suspicious receiver, suspicious sensors, active eavesdropper, and intelligent reflecting surface in the system as Figure 1 shown, construct an active eavesdropping model of the suspicious air computing system.

[0081] Step 202, according to the active eavesdropping model of the suspicious air computing system, obtain the computational mean square error of the suspicious receiver when collecting illegal information through the suspicious sensors by the transmit power of the suspicious sensors, the receive power of the suspicious receiver, and the phase matrix of the intelligent reflecting surface.

[0082] Among them, the calculated mean square error when the suspicious receiver collects illegal information through the suspicious sensors is the same as the calculated mean square error when the suspicious receiver aggregates the illegal information collected by K suspicious sensors through air computing. Among them, the suspicious receiver and the suspicious sensors are users in the legitimate network who have been bribed or hijacked by the adversary. Since they are users in the legitimate network, they can obtain the transmission power of the suspicious sensors and the reception power of the suspicious receiver.

[0083] Step 203: According to the active listening model of the suspicious air computing system, obtain the calculated mean square error of the active listener when listening to the suspicious receiver collecting illegal information through the reception power of the active listener, the active interference power of the active listener, and the phase matrix of the intelligent reflecting surface.

[0084] In Steps 202 and 203, the calculated mean square error when the suspicious receiver collects illegal information through the suspicious sensors is expressed by the following formula:

[0085]

[0086] The calculated mean square error of the active listener when listening to the suspicious receiver collecting illegal information is expressed by the following formula:

[0087]

[0088] In the formula, represents the signal misalignment error of the suspicious receiver, represents the error caused by the active interference of the active listener, represents the error caused by the noise of the suspicious receiver, represents the signal misalignment error of the active listener, represents the error caused by the full-duplex self-interference of the active listener, represents the error caused by the noise of the active listener, Φ = diag(φ1, φ2, …, φ N ) represents the phase matrix of the intelligent reflecting surface, n ∈ {1, 2, …, N}, θ n ∈ [0, 2π), represents the adjusted phase of the nth reflection unit of the intelligent reflecting surface, u D and u L respectively represent the reception powers of the suspicious receiver and the active listener, v k represents the transmission power of the kth suspicious sensor, w represents the active interference power of the active listener, and respectively represent the noise powers of the suspicious receiver and the active eavesdropper, and respectively represent the channels from the k-th suspicious sensor to the suspicious receiver, the intelligent reflecting surface, and the active eavesdropper, \(H\) ID and \(H\) IL respectively represent the channels from the intelligent reflecting surface to the suspicious receiver and the active eavesdropper, \(H\) LD represents the channel from the active eavesdropper to the suspicious receiver, \(H\) LL represents the full-duplex self-interference channel of the active eavesdropper.

[0089] Step 204, under the constraints of the calculated mean square error of the suspicious receiver, the transmission power of the suspicious sensor, the active interference power of the active eavesdropper, and the phase matrix of the intelligent reflecting surface, construct an optimization problem with the goal of minimizing the mean square error of the active eavesdropper.

[0090] Among them, the optimization problem with the goal of minimizing the mean square error of the active eavesdropper is expressed by the following formula:

[0091]

[0092] In the formula, \(MSE\) L represents the calculated mean square error of the active eavesdropper, \(C1\) represents the constraint condition of the calculated mean square error of the suspicious receiver, \(\varepsilon\) represents the preset covert constraint threshold, which is a constant, \(C2\) represents the constraint condition of the transmission power of the suspicious sensor, \(P\) k represents the maximum transmission power of the k-th suspicious sensor, \(C3\) represents the constraint condition of the active interference power of the active eavesdropper, \(P\) L represents the maximum transmission power of the active eavesdropper, \(C4\) represents the constraint condition of the phase matrix of the intelligent reflecting surface, \(n\in\{1,2,\cdots,N\}\) represents the n-th reflection unit of the intelligent reflecting surface.

[0093] In the optimization problem \(P1\), the objective function and each optimization variable in the constraint \(C1\) are highly coupled, and the constraint \(C4\) is non-convex. Therefore, the optimization problem \(P1\) is a non-convex problem, and the Sequential Rank-One Constraint Relaxation (SROCR) iterative algorithm can be used to solve \(P1\).

[0094] Step 205, solve the optimization problem with the goal of minimizing the calculated mean square error of the active eavesdropper to obtain the target received power of the active eavesdropper, the target active interference power of the active eavesdropper, and the target phase matrix of the intelligent reflecting surface.

[0095] Specifically, first, the optimization problem aiming at minimizing the computational mean square error of the active listener is decomposed into multiple sub-optimization problems, each aiming at minimizing the computational mean square error of the active listener and taking the received power of the suspicious receiver and the active listener, the transmission power of the suspicious sensor, the active interference power of the active listener, and the phase matrix of the intelligent reflecting surface as optimization variables. Then, each sub-optimization problem is solved.

[0096] Among them, the sub-optimization problem of minimizing the received power of the suspicious receiver and the active listener with the goal of minimizing the computational mean square error of the active listener is expressed by the following formula:

[0097] P2:

[0098] s.t.C1:

[0099] The sub-optimization problem of minimizing the transmission power of the suspicious sensor with the goal of minimizing the computational mean square error of the active listener is expressed by the following formula:

[0100] P3:

[0101] s.t.C2:

[0102] The sub-optimization problem of minimizing the active interference power of the active listener with the goal of minimizing the computational mean square error of the active listener is expressed by the following formula:

[0103] P5:

[0104] s.t.C1:

[0105] C3: ||w|| ≤ P

[0106] The sub-optimization problem of minimizing the phase matrix of the intelligent reflecting surface with the goal of minimizing the computational mean square error of the active listener is expressed by the following formula:

[0107] P6:

[0108] s.t.C1:

[0109] C4:

[0110] In the specific implementation, for the sub-optimization problem P2, when optimizing the received power {u D, u L} of the suspicious receiver and the active listener, by setting the transmission power {v k}, the active interference power w of the active eavesdropper, and the phase matrix Φ of the intelligent reflecting surface. The sub-optimization problem P2 is an unconstrained optimization problem of {u D, u L}. The closed-form expressions of the optimal solutions of the received powers of the legitimate receiver and the active eavesdropper are given by the following formulas:

[0111]

[0112] where, denotes the equivalent channel from the k-th legitimate sensor to the legitimate receiver, denotes the equivalent channel from the k-th legitimate sensor to the active eavesdropper.

[0113] For the sub-optimization problem P3, when optimizing the transmit power {v k} of the legitimate sensors, by given the received powers {u D, u L} of the legitimate receiver and the active eavesdropper, the active interference power w of the active eavesdropper, and the phase matrix Φ of the intelligent reflecting surface, the sub-optimization problem P3 can be further decoupled into K sub-sub-optimization problems P4 that can be executed in parallel as follows:

[0114] P4:

[0115] s.t. C2: |v k | 2 ≤ P k

[0116] By the sub-sub-optimization problem, using the Lagrange multiplier method, the closed-form expression of the optimal solution of the transmit power of the legitimate sensors is obtained as the following formula:

[0117]

[0118] where, denotes the Lagrange multiplier corresponding to the constraint condition C2.

[0119] For the sub-optimization problem P5, when optimizing the active interference power w of the active eavesdropper, by given the received powers {u D , u L} of the legitimate receiver and the active eavesdropper, the transmit power {v k} of the legitimate sensors, and the phase matrix Φ of the intelligent reflecting surface, the sub-optimization problem of minimizing the active interference power of the active eavesdropper is transformed into a generalized Rayleigh quotient problem. By the generalized Rayleigh quotient problem, the closed-form expression of the optimal solution of the active interference power of the active eavesdropper is obtained as the following formula:

[0120]

[0121] where, denotes the eigenvector corresponding to the minimum eigenvalue of matrix (A D / ε1) -1 A L .

[0122] When optimizing the phase matrix Φ of the intelligent reflecting surface for the sub-optimization problem P6, by given the received powers {u D , u L} of the suspicious receivers and the active listeners, the transmitted powers {v k} of the suspicious sensors, and the active interference power w of the active listeners, using the identity tr(SFTF H ) = f H (S⊙T T )f, and introducing the preset auxiliary variable t, the optimization problem P6 can be equivalently expressed as:

[0123] P7:

[0124] s.t. C1:

[0125] C4:

[0126] where F = diag(f) represents the diagonal matrix with f as the diagonal elements, t represents the auxiliary variable,

[0127] c D = ([C D 1,1 , [C D 2,2 , …, [C D N,N ), c L = ([C L 1,1 , [C L 2,2 , …, [C L N,N ), [A] n,n denotes the n-th diagonal element of matrix A, obtained by performing eigen-decomposition on the optimal solution X opt of P8, and g max is the eigenvector corresponding to the maximum eigenvalue λ opt of X max .

[0128] ​​​​​Then, by introducing a variable the problem after equivalent transformation is again transformed into the following rank-1 positive definite programming problem:

[0129] P8:

[0130] s.t. C1: tr(R D X) ≥ ε2

[0131] C5:

[0132] C6: X ≥ 0

[0133] C7: rank(X) = 1

[0134] P8 is a rank-1 positive definite programming problem, and the alternating continuous rank-one constraint relaxation iterative algorithm can be used to solve it. After obtaining the optimal solution X opt of problem P8, perform eigenvalue decomposition on X opt ; g max is the eigenvector corresponding to the largest eigenvalue λ opt of X max ; then the closed-form expression of the optimal solution of the phase matrix of the intelligent reflecting surface is

[0135] Based on the above, for each sub-optimization problem, the alternating continuous rank-one constraint relaxation iterative SROCR algorithm is used to solve it, which specifically includes the following steps:

[0136] (1) Set the initial value of the calculated mean square error of the active listener , the initial value of the iteration number i = 1, the iteration accuracy δ, and the maximum iteration number I max , and set the initial values of each optimization variable; among them, the optimization variables include the transmission power of the suspicious sensor , the active interference power w (0) of the active listener, and the initial value Φ (0) of the phase matrix of the intelligent reflecting surface;

[0137] (2) Based on the values Φ (i-1) and w (i-1) of the optimization variables in the previous iteration, respectively calculate the optimal received power of the suspicious receiver and the optimal received power

[0138] (3) Based on the values Φ (i-1) , w (i-1) of the optimization variables in the previous iteration, Obtain the transmission power of the optimal suspicious sensor according to the closed-form expression of the optimal solution of the transmission power of the suspicious sensor

[0139] (4) Based on the value of the optimization variable Φ in the previous iteration (i-1) , Obtain the active interference power w of the optimal active listener according to the closed-form expression of the optimal solution of the active interference power of the active listener (i) ;

[0140] (5) Based on the value of the optimization variable in the previous iteration w (i) , obtain the phase matrix Φ of the optimal intelligent reflecting surface according to the closed-form expression of the optimal solution of the phase matrix of the intelligent reflecting surface (i) ;

[0141] (6) Increase the iteration number i = i + 1;

[0142] (7) If i > I max or then stop the iteration.

[0143] Step 206: Make the active listener monitor the suspicious receiver to collect illegal information with the target receiving power, and actively interfere with the suspicious receiver to collect illegal information with the target active interference power. Make the intelligent reflecting surface enhance the monitoring channel of the active listener and weaken the suspicious channel of the suspicious receiver with the target phase matrix, so as to actively monitor the suspicious receiver in the suspicious air computing system.

[0144] In one example, the following simulation verification is carried out for the active monitoring method of the suspicious air computing system of this embodiment:

[0145] The simulation parameters are set as follows: N = 30, M D = 6, M T = M R = 4, δ = 0.001, I max = 200, ε = K, all suspicious sensors have the same maximum transmission power constraint, that is, P1 = P2 = … = P K = 10dBm, P L = 10dBm. All suspicious sensors are evenly distributed in a circle with the center at (0, 0, 0) and a radius of 2. The coordinates of the suspicious receiver are (10, 0, 0), the coordinates of the IRS are (1, 1, 2), and the coordinates of the active listener are (2, 1, 2). See Figure 1 , all channels follow a circularly complex Gaussian random distribution with a mean of 0 and a variance of , d ij , i ∈ {sk , where \(d_{ij}\), \(i,j\in\{D, I, L\}\), is the distance between node \(i\) and node \(j\), and \(\alpha\) ij is the corresponding path loss factor. \(\alpha\) IL \(=\alpha\) ID \(=\alpha\) LD \( = 2.8\). MATLAB software is used for simulation, and the number of Monte Carlo simulations is 200 times.

[0146] Figure 3 This is a comparison graph of the mean square error calculated by the method of this embodiment and other methods. It can be found from the graph that the performance of the mean square error calculated by all algorithms deteriorates as the number of sensors \(K\) increases. This is because as the number of sensors \(K\) increases, it becomes more difficult for the active listener to design a receiving beam that can aggregate all sensor information. It can also be found from the graph that the method of this embodiment is far superior to the algorithms without IRS and with random IRS phases. This is because by optimizing the IRS phase, the received signal of the active listener can be enhanced while that of the suspicious receiver can be weakened, achieving the purpose of improving the performance of the calculated mean square error. In addition, the improvement of the performance of the method of this embodiment for the passive listening algorithm is limited. This is because while the active interference weakens the received signal of the suspicious receiver, it also generates self-interference to the active listener, limiting the degree of performance improvement of the listening system.

[0147] In this embodiment, by minimizing the mean square error of the active listener when listening to and interfering with the suspicious receiver collecting illegal information, the success rate of active listening of the air computing system can be improved. Therefore, the present invention constructs an optimization problem with the goal of minimizing the mean square error of the active listener and solves it to obtain the operating parameters of each device in the system when the mean square error of the active listener is minimized, such as the receiving power of the active listener, the active interference power of the active listener, and the phase matrix of the intelligent reflecting surface. Thus, with their respective operating parameters cooperating with each other, the active listening of the suspicious air computing system can be realized. It includes the operating parameters of the intelligent reflecting surface, which can be used to enhance the listening channel for the active listener to listen to and actively interfere with the suspicious receiver collecting illegal information and weaken the suspicious channel for the suspicious receiver to collect illegal information. Therefore, by setting an intelligent reflecting surface in the suspicious air computing system, the mean square error of the suspicious receiver when collecting illegal information through suspicious sensors can be increased. On this basis, by minimizing the mean square error of the active listener when listening to and interfering with the suspicious receiver collecting illegal information, the success rate of active listening is further improved.

[0148] The step division of the above various methods is only for clear description. When implementing, they can be combined into one step or some steps can be split into multiple steps. As long as the same logical relationship is included, it is within the protection scope of the present invention; adding insignificant modifications to the algorithm or process or introducing insignificant designs, but without changing the core design of the algorithm and process, is within the protection scope of the invention.

[0149] Another embodiment of the present invention relates to an active monitoring system for a suspicious air computing system. The suspicious air computing system includes a suspicious receiver and a suspicious sensor. The active monitoring system includes an active monitor and an intelligent reflecting surface, and there is a connection relationship between any two of the suspicious receiver, the suspicious sensor, the active monitor, and the intelligent reflecting surface; wherein, the suspicious receiver is used to collect illegal information through the suspicious sensor, the active monitor is used to monitor and actively interfere with the suspicious receiver's collection of illegal information, and the active monitor has self-interference. The intelligent reflecting surface is used to enhance the monitoring channel for the active monitor to monitor the suspicious receiver's collection of illegal information and weaken the suspicious channel for the suspicious receiver to collect illegal information.

[0150] The following specifically describes the implementation details of the active monitoring system for the suspicious air computing system in this embodiment. The following content is only the implementation details provided for convenient understanding and is not necessary for implementing this solution.

[0151] The active monitoring system for the suspicious air computing system in this embodiment further includes:

[0152] A monitoring model construction module, which is used to construct an active monitoring model of the suspicious air computing system according to the channels between the suspicious sensor and the suspicious receiver, the channels between the intelligent reflecting surface and the suspicious receiver, the channels between the suspicious sensor and the active monitor, the channels between the suspicious sensor, the intelligent reflecting surface and the active monitor, the channels between the active monitor and the suspicious receiver, and the self-interference channel of the active monitor;

[0153] A first data acquisition module, which is used to obtain the calculated mean square error of the suspicious receiver when collecting illegal information through the suspicious sensor according to the active monitoring model of the suspicious air computing system, the transmission power of the suspicious sensor, the reception power of the suspicious receiver, and the phase matrix of the intelligent reflecting surface;

[0154] A second data acquisition module, which is used to obtain the calculated mean square error of the active monitor when monitoring the suspicious receiver's collection of illegal information according to the active monitoring model of the suspicious air computing system, the reception power of the active monitor, the active interference power of the active monitor, and the phase matrix of the intelligent reflecting surface;

[0155] An optimization problem construction module, configured to construct an optimization problem with the goal of minimizing the calculated mean square error of the active listener under the constraints of the calculated mean square error of the suspicious receiver, the transmission power of the suspicious sensor, the active interference power of the active listener, and the phase matrix of the intelligent reflecting surface;

[0156] An optimization problem solving module, configured to solve the optimization problem with the goal of minimizing the mean square error of the active listener, and obtain the target receiving power of the active listener, the target active interference power of the active listener, and the target phase matrix of the intelligent reflecting surface;

[0157] An active listening module, configured to enable the active listener to listen for illegal information collected by the suspicious receiver with the target receiving power, and actively interfere with the suspicious receiver to collect illegal information with the target active interference power, so that the intelligent reflecting surface enhances the listening channel of the active listener and weakens the suspicious channel of the suspicious receiver with the target phase matrix, so as to actively listen to the suspicious receiver in the suspicious air computing system.

[0158] It is not difficult to find that this embodiment is a system embodiment corresponding to the above method embodiment, and this embodiment can be implemented in cooperation with the above method embodiment. The relevant technical details and technical effects mentioned in the above embodiments are still valid in this embodiment. For the sake of reducing repetition, they will not be elaborated here. Correspondingly, the relevant technical details mentioned in this embodiment can also be applied to the above embodiments.

[0159] It is worth mentioning that each module involved in this embodiment is a logical module. In practical applications, a logical unit can be a physical unit, a part of a physical unit, or a combination of multiple physical units. In addition, in order to highlight the innovative part of the present invention, units that are not closely related to solving the technical problems proposed by the present invention are not introduced in this embodiment, but this does not mean that there are no other units in this embodiment.

[0160] Another embodiment of the present invention relates to a computer device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor, so that the at least one processor can execute the active listening method of the suspicious air computing system in the above embodiments.

[0161] Among them, the memory and the processor are connected in a bus manner. The bus may include any number of interconnected buses and bridges, which connect various circuits of one or more processors and the memory together. The bus may also connect various other circuits, such as peripheral devices, voltage regulators, and power management circuits, etc., which are well known in the art. Therefore, they will not be further described herein. The bus interface provides an interface between the bus and the transceiver. The transceiver may be an element or multiple elements, such as multiple receivers and transmitters, and provides a unit for communicating with various other devices on the transmission medium. The data processed by the processor is transmitted on the wireless medium through the antenna. Further, the antenna also receives data and transmits the data to the processor.

[0162] The processor is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interface, voltage regulation, power management, and other control functions. The memory can be used to store the data used by the processor when executing operations.

[0163] Another embodiment of the present invention relates to a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, the method embodiments described above are implemented.

[0164] That is, those skilled in the art can understand that all or part of the steps in implementing the methods of the above embodiments can be completed by instructing relevant hardware through a program. The program is stored in a storage medium, including several instructions to enable a device (which can be a single-chip microcomputer, a chip, etc.) or a processor to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs, etc., which can store program codes.

[0165] Those of ordinary skill in the art can understand that the above embodiments are specific embodiments for implementing the present invention, and in practical applications, various changes can be made in form and details without departing from the spirit and scope of the present invention.

Claims

1. An active monitoring method for a suspicious air computing system, characterized in that, Applied to an active monitoring system, the suspicious air computing system includes a suspicious receiver and a suspicious sensor. The active monitoring system includes an active monitor and an intelligent reflecting surface, and there is a connection relationship between any two of the suspicious receiver, the suspicious sensor, the active monitor, and the intelligent reflecting surface; Among them, the suspicious receiver is used to collect illegal information through the suspicious sensor. The active monitor is used to monitor and actively interfere with the suspicious receiver's collection of illegal information, and the active monitor has self-interference. The intelligent reflecting surface is used to enhance the monitoring channel for the active monitor to monitor the suspicious receiver's collection of illegal information and weaken the suspicious channel for the suspicious receiver to collect illegal information; The method includes: Construct an active monitoring model of the suspicious air computing system according to the channel between the suspicious sensor and the suspicious receiver, the channel between the intelligent reflecting surface and the suspicious receiver, the channel between the suspicious sensor and the active monitor, the channel between the suspicious sensor, the intelligent reflecting surface and the active monitor, the channel between the active monitor and the suspicious receiver, and the self-interference channel of the active monitor; According to the active monitoring model of the suspicious air computing system, obtain the computational mean square error of the suspicious receiver when collecting illegal information through the suspicious sensor by the transmit power of the suspicious sensor, the receive power of the suspicious receiver, and the phase matrix of the intelligent reflecting surface; According to the active monitoring model of the suspicious air computing system, obtain the computational mean square error of the active monitor when monitoring the suspicious receiver's collection of illegal information by the receive power of the active monitor, the active interference power of the active monitor, and the phase matrix of the intelligent reflecting surface; Under the constraints of the computational mean square error of the suspicious receiver, the transmit power of the suspicious sensor, the active interference power of the active monitor, and the phase matrix of the intelligent reflecting surface, construct an optimization problem with the goal of minimizing the computational mean square error of the active monitor; Solve the optimization problem with the goal of minimizing the computational mean square error of the active monitor to obtain the target receive power of the active monitor, the target active interference power of the active monitor, and the target phase matrix of the intelligent reflecting surface; Enable the active monitor to monitor the suspicious receiver's collection of illegal information with the target receive power and actively interfere with the suspicious receiver's collection of illegal information with the target active interference power. Enable the intelligent reflecting surface to enhance the monitoring channel of the active monitor and weaken the suspicious channel of the suspicious receiver with the target phase matrix to actively monitor the suspicious air computing system.

2. The active listening method of the suspicious air computing system according to claim 1, characterized in that, The solution to the optimization problem with the goal of minimizing the computational mean square error of the active monitor includes: Decompose the optimization problem with the goal of minimizing the computational mean square error of the active monitor to obtain multiple sub-optimization problems with the goal of minimizing the computational mean square error of the active monitor, and with the receive power of the suspicious receiver and the active monitor, the transmit power of the suspicious sensor, the active interference power of the active monitor, and the phase matrix of the intelligent reflecting surface as optimization variables respectively; Solve each sub-optimization problem.

3. The active listening method of the suspicious air computing system according to claim 2, characterized in that, The solution to each sub-optimization problem includes: For each sub-optimization problem, use the alternating successive rank-one constraint relaxation iteration SROCR algorithm to solve it.

4. The active listening method of the suspicious air computing system according to claim 3, characterized in that, The mean square error of the calculation when the suspected receiver collects illegal information through the suspected sensor is expressed by the following formula: The mean square error of the calculation when the active eavesdropper eavesdrops on the suspected receiver collecting illegal information is expressed by the following formula: wherein, represents the signal misalignment error of the suspected receiver, represents the error caused by the active interference of the active eavesdropper, represents the error caused by the noise of the suspected receiver, represents the signal misalignment error of the active eavesdropper, represents the error caused by the full-duplex self-interference of the active eavesdropper, represents the error caused by the noise of the active eavesdropper, Φ = diag(φ1, φ2, …, φ N ) represents the phase matrix of the intelligent reflecting surface, n ∈ {1, 2, …, N}, θ n ∈ [0, 2π), represents the adjusted phase of the nth reflecting element of the intelligent reflecting surface, u D and u L respectively represent the received powers of the suspected receiver and the active eavesdropper, v k represents the transmission power of the kth suspected sensor, w represents the active interference power of the active eavesdropper, and respectively represent the noise powers of the suspected receiver and the active eavesdropper, and respectively represent the channels from the kth suspected sensor to the suspected receiver, the intelligent reflecting surface and the active eavesdropper, H ID and H IL respectively represent the channels from the intelligent reflecting surface to the suspected receiver and the active eavesdropper, H LD represents the channel from the active eavesdropper to the suspected receiver, H LL represents the full-duplex self-interference channel of the active eavesdropper.

5. The active listening method of the suspicious air computing system according to claim 4, characterized in that, The optimization problem aiming at minimizing the mean square error of the calculation of the active eavesdropper is expressed by the following formula: where, MSE L represents the calculated mean square error of the active eavesdropper, C1 represents the constraint condition of the calculated mean square error of the suspicious receiver, ε represents the preset covert constraint threshold, which is a constant, C2 represents the constraint condition of the transmission power of the suspicious sensor, P k represents the maximum transmission power of the k-th suspicious sensor, C3 represents the constraint condition of the active interference power of the active eavesdropper, P L represents the maximum transmission power of the active eavesdropper, C4 represents the constraint condition of the phase matrix of the intelligent reflecting surface, and n ∈ {1, 2, …, N} represents the n-th reflecting element of the intelligent reflecting surface.

6. The active listening method of the suspicious air computing system according to claim 5, characterized in that, Before solving each sub-optimization problem, it further includes: By given the transmit power of the suspected sensor, the active interference power of the active eavesdropper, and the phase matrix of the intelligent reflecting surface, the closed-form expressions of the optimal solutions of the received powers of the suspected receiver and the active eavesdropper are obtained, as shown in the following formula: wherein, represents the equivalent channel from the k-th suspicious sensor to the suspicious receiver, represents the equivalent channel from the k-th suspicious sensor to the active eavesdropper; By given the received powers of the suspected receiver and the active eavesdropper, the active interference power of the active eavesdropper, and the phase matrix of the intelligent reflecting surface, using the Lagrange multiplier method, the closed-form expression of the optimal solution of the transmit power of the suspected sensor is obtained, as shown in the following formula: wherein, represents the Lagrange multiplier corresponding to the constraint condition C2; By given the received powers of the suspected receiver and the active eavesdropper, the transmit power of the suspected sensor, and the phase matrix of the intelligent reflecting surface, the closed-form expression of the optimal solution of the active interference power of the active eavesdropper is obtained, as shown in the following formula: In the formula, represents the eigenvector corresponding to the minimum eigenvalue of the matrix (A D / ε1) -1 A L , By given the received powers of the suspected receiver and the active eavesdropper, the transmit power of the suspected sensor, and the active interference power of the active eavesdropper, the closed-form expression of the optimal solution of the phase matrix of the intelligent reflecting surface is obtained, as shown in the following formula: where \(F = \text{diag}(f)\) represents a diagonal matrix with \(f\) as the diagonal elements, and \(t\) represents an auxiliary variable. c D =([C D 1,1 ,[C D 2,2 ,…,[C D N,N ), \(c\) L =([C L 1,1 ,[C L 2,2 ,…,[C L N,N ), \([A] n,n represents the \(n\)-th diagonal element of matrix \(A\). is obtained by performing eigen - decomposition on the optimal solution \(X\) of \(P8\). \(g\) opt is the eigen - vector corresponding to the largest eigenvalue \(\lambda\) max of \(X\) opt max .​​​​​​ 7. The active listening method of the suspicious air computing system according to claim 6, characterized in that For each sub-optimization problem, the alternating sequential rank-one constraint relaxation iterative SROCR algorithm is used for solving, including: Set the initial value of the calculated mean square error of the active listener The initial value of the iteration count \(i = 1\), the iteration precision \(\delta\), and the maximum iteration count \(I\) max , and set the initial values of each optimization variable; the optimization variables include the transmission power of the suspicious sensor The active interference power \(w\) of the active listener (0) and the initial value \(\varPhi\) of the phase matrix of the intelligent reflecting surface (0) ; Based on the values of the optimized variables in the previous iteration According to the closed-form expressions of the optimal solutions of the received powers of the suspicious receiver and the active eavesdropper, respectively, the optimal received power of the suspicious receiver is obtained and the optimal received power of the active eavesdropper Based on the values of the optimized variables in the previous iteration Obtain the optimal transmission power of the suspicious sensor according to the closed-form expression of the optimal solution of the transmission power of the suspicious sensor Based on the values of the optimized variables in the previous iteration Obtain the optimal active interference power w of the active listener according to the closed expression of the optimal solution of the active interference power of the active listener (i) ; Based on the values of the optimized variables in the previous iteration Obtain the optimal phase matrix Φ of the intelligent reflecting surface according to the closed-form expression of the optimal solution of the phase matrix of the intelligent reflecting surface (i) ; Increase the iteration number i = i + 1; If i > I max or , stop the iteration.

8. An active monitoring system for a suspicious air computing system, characterized in that, The suspected air computing system includes a suspected receiver and a suspected sensor, the active eavesdropping system includes an active eavesdropper and an intelligent reflecting surface, and there is a connection relationship between any two of the suspected receiver, the suspected sensor, the active eavesdropper, and the intelligent reflecting surface; Among them, the suspected receiver is used to collect illegal information through the suspected sensor, the active eavesdropper is used to eavesdrop on and actively interfere with the suspected receiver collecting illegal information, and the active eavesdropper has self-interference. The intelligent reflecting surface is used to enhance the eavesdropping channel for the active eavesdropper to eavesdrop on the suspected receiver collecting illegal information and weaken the suspected channel for the suspected receiver to collect illegal information; The active eavesdropping system further includes: An eavesdropping model construction module, which is used to construct an active eavesdropping model of the suspected air computing system according to the channels between the suspected sensor and the suspected receiver, the channels between the intelligent reflecting surface and the suspected receiver, the channels between the suspected sensor and the active eavesdropper, the channels between the suspected sensor, the intelligent reflecting surface and the active eavesdropper, the channels between the active eavesdropper and the suspected receiver, and the self-interference channel of the active eavesdropper; A first data acquisition module, which is used to obtain the mean square error of the calculation when the suspected receiver collects illegal information through the suspected sensor according to the active eavesdropping model of the suspected air computing system, through the transmit power of the suspected sensor, the received power of the suspected receiver, and the phase matrix of the intelligent reflecting surface; The second data acquisition module is configured to obtain the calculated mean square error of the active listener when listening to the illegal information collected by the suspicious receiver according to the active listening model of the suspicious air computing system, through the received power of the active listener, the active interference power of the active listener, and the phase matrix of the intelligent reflecting surface; The optimization problem construction module is configured to construct an optimization problem with the goal of minimizing the calculated mean square error of the active listener under the constraints of the calculated mean square error of the suspicious receiver, the transmission power of the suspicious sensor, the active interference power of the active listener, and the phase matrix of the intelligent reflecting surface; The optimization problem solving module is configured to solve the optimization problem with the goal of minimizing the mean square error of the active listener, and obtain the target received power of the active listener, the target active interference power of the active listener, and the target phase matrix of the intelligent reflecting surface; The active listening module is configured to enable the active listener to listen to the illegal information collected by the suspicious receiver with the target received power, and actively interfere with the illegal information collected by the suspicious receiver with the target active interference power, and enable the intelligent reflecting surface to enhance the listening channel of the active listener and weaken the suspicious channel of the suspicious receiver with the target phase matrix, so as to actively listen to the suspicious receiver in the suspicious air computing system.

9. A computer device, characterized in that, Comprising: At least one processor; And a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor, so that the at least one processor can execute the active listening method of the suspicious air computing system according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the active listening method of the suspicious air computing system according to any one of claims 1 to 7.

Images